Scan failed

Understand pipeline failures across crawl, analyze, and probe — and recover with a clean retry.

Updated 2026-08-06

A scan is marked failed when a hard step cannot finish. VisibleMax runs crawl → analyze → probe under Trigger.dev. Crawl and core analysis must succeed for a usable report; probe enrichment can soft-fail without wiping the report.

Hard failure vs soft failure

StageFailure modeWhat you see
CrawlHardScan failed — no pages fetched or site unreachable
AnalyzeHard (core)Scan/report failed — analysis could not persist
Probe / discoverySoftStatus partial or completed with fewer platforms

Common causes

  • Unreachable site — DNS NXDOMAIN, connection refused, or HTTP errors on the homepage
  • SSL / certificate problems — expired, mismatched, or incomplete chains
  • Bot blocking / WAF — Cloudflare, firewall, or bot challenges that block crawlers
  • Timeouts — crawl or analysis exceeded time budgets on large or slow sites
  • AI provider key missing (probe only) — platforms skip; usually not a full scan failure
  • Free plan already used — lifetime completed scan consumed; start is blocked until Pro

Recovery steps

  1. 1

    Read the scan error

    Open the scan detail page and copy error_message / status. Match it in Common errors.

  2. 2

    Verify the URL

    Confirm the project URL uses the correct scheme (https://), host, and path. Fix redirects that loop or land on auth walls.

  3. 3

    Fix reachability

    Resolve DNS/SSL and allowlist crawler traffic if a WAF is blocking. Details: Website unreachable.

  4. 4

    Check plan gates

    If the UI asks you to upgrade, your Free lifetime scan is used. Upgrade to VisibleMax Pro — Early Adopter (7-day free trial, then $50/year) via Polar, or contact support if that is wrong.

  5. 5

    Retry once

    Start a new scan after fixes. Avoid hammering retries — Trigger.dev already retries transient steps.

  6. 6

    Escalate with evidence

    If it fails again with a public site, email support with project URL, scan ID, steps tried, and screenshots.

Edge cases

  • Competitor / core mode — some jobs run crawl → analyze only (no probe). A missing probe section is expected, not a failure.
  • Stuck in queued/running — may be orchestration delay, not failure yet. Wait a few minutes; see Slow scans.
  • Failed after pages crawled — analyze-side issue; include scan ID and approximate page count when contacting support.
  • Auth-only sites — public marketing pages scan; gated app shells often yield thin or failed crawls.

When to contact support

  • Two consecutive failures after reachability is confirmed
  • Error message is empty or generic with no matching doc
  • Failure coincides with a billing change you already completed

FAQ

Can I resume a failed scan mid-pipeline?

Start a new scan after fixing the cause. Durable orchestration retries steps automatically, but a hard-failed scan should be replaced with a clean run.

Why did probe fail but I still have a report?

Probe soft-fails. Core analyze unlocks the report first; probe enriches in parallel. Partial coverage is expected when providers skip.

Will retrying burn my Free scan?

Free allows one completed scan lifetime. Hard failures generally leave the allowance available; completed Free scans require Pro for more.

Related

Still stuck? Contact support with your project URL and what you already tried.